Forecasting Customer Decisions, Based on Digital Footprint in Facebook for a Company From the Cosmetics Sector

This work was done to find the optimal method for predicting client's choice based on digital traces on social networks. The purpose of this dissertation is to identify the best practices for predicting customer digital footprint for M on Facebook by analyzing a cosmetics company's 2020 ad campaigns. The theoretical part of the work is aimed at studying relevant information using methods that predict client's choice based on a digital footprint. It consists of the toric foundations of the digital footprint, methods, technology and tools for analyzing digital footprints. The technologies and algorithms of Facebook were described in more detail, and the launch of advertising based on data on the VK site was also considered for comparison. In addition, books on marketing and several online articles were studied, mainly foreign experience was studied. The dissertation also covers how to see digital user data and measure the results of ad campaigns that have been customized based on user digital footprints using Facebook Insights, Facebook Ads Manager, and Facebook Analytics. Then the theory is applied already in the practical part, where an in-depth analysis of the results of advertising campaigns using the Python programming language is carried out to identify patterns of behavior and understand the choice of a company product from the "cosmetics" sector based on their traces in advertising campaigns on social networks. The research result is applied to existing campaigns to measure the effectiveness and a conclusion is drawn based on the campaigns completed.